insullation beneath floor with radiant heat

CMadatMe

The more r-value you have below the less inflitration you get. Sounds like your floor r-value is around a 1.5. As long as my r-value below is higher then the 1.5 you will get heat transfer. Min insulation r-value for joist heating is a the least an R-11. If your using Wirsbo's joist trak plates you insullate directly to the plate. The reason other mfs have the spacing is that they need to create convection in the joist bay to get the btu output through the floor. With joist trak plates you are actually transfering the heat absorbed by the plate directly to the floor not creating an oven in the joist bay.

Lance Macey

insullation under floor with radiant heat

I have true radiant heat installed beneath my 4 inch width white oak floor. The system consists of plates nailed directly to the subfloor with the pex tubing snapped into the plates. My heating contractor's goal is to run as low a temperature water throught the system as possible to heat the floor and not cause cupping and crowning that can result from over heating and drying out the wood. The basement environment is an unheated (New England cold) crawl space. The question is: with icynene foam insulation in place, am I better off with the spaces between the floor joists being completely filled so that the foam is in direct contact with the radiant tubes and plates OR is it more effecient to have a 2 inch air space between the radiant system and the foam insulation? No one seems to be able to answer this question with a good thermodynamic explanation. Does it depend upon the heat absorption characteristics of the foam? Is one way better than the other? Thank you for your help

Gordan

That doesn't quite add up, theoretically.

(Disclaimer: I'm not a heating professional.)



It would seem to me that, as long as you're applying the same amount of insulation, applying it lower - and not directly to the bottom of the plates - would decrease the heat flow. Since the top of the cavity is hotter than the bottom, convection between the joists should not be an issue, meaning that you'd benefit from the insulating value of "dead" air above the insulation itself - R5 per inch or so.



The plates themselves have low emissivity, so heat flow through radiation should also be low (except for areas of the subfloor which are not covered by plates) even without an upward-facing radiant barrier (which could be subject to reduced performance through dust contamination.)



If this is an option, insulating below the joists with something that would prevent air infiltration (e.g. rigid board) would seem to provide the most benefit for your money, as it would also prevent thermal bridging through the joists.

Jamie_5

insulation air space

I think leaving the air gap would increase the thermal bridging at the joist and be less efficient.  I agree that insulating below the joists would be best, if practical.

Gordan

Bridging through joists

Jamie, the whole purpose of the heat transfer plate is to conduct to the subfloor. If the subfloor conducts to the joists, that'll present less resistance to heat flow than conduction through the air.

Jamie_5

air gap as insulator

Please bear with me, as I may simply be dense.  I understand that the purpose of the transfer plate is to use conduction and increase (over a naked tube) the rate of heat flow into the subfloor.  My response was only meant to question your suggestion that the air gap would add insulating value below the plates.  I thought that by leaving the air gap and thereby exposing more surface area of the joist to warm air in the cavity, you would increase the rate of heat transfer through the joists.  Perhaps what I was not considering---because I didn't know---was your point about the emissivity of the plate.  Perhaps this means that the plate warms the joist cavity very little relative to its conduction into the subfloor.  I suppose if we knew the rate of heat transfer into the  empty cavity (if you had the air gap), we could calculate whether leaving the air gap was a net win.

Gordan

Something along those lines...

Heat through a fluid moves through three means: conduction, radiation and convection. In this case, convection would be unlikely to occur, leaving radiation and conduction. Radiation from the plates would be minimized by their own low emissivity. So you're left with conduction through the subfloor and joists, conduction through the air, and radiation from the areas of the subfloor not covered by the plates.



The conduction path through the subfloor to the joists, and then through the joist to the lowest exposed (to the cavity) point on the joist, is a little longer than the conduction path through the air to the same point on the joist (Pythagoras' theorem :-)) but the R-value of dead air is about three times that of wood. This may mean that the joists would actually be warmer than the air next to them, which would mean that they would heat, rather than cool, that air. (If you were still worried about some sort of short-circuit, you could put foil-faced polyiso - not much thickness would be needed here - on the exposed joist sides, and that could act as a convenient "furring strip" for the lower insulation.)



That's theory. In practice, I've been searching for real world studies and haven't been able to find any. For what it's worth, Viega recommends a 1/2" gap under their plates (OEM-d ThinFin, I think) if no air infiltration exists. 1/2" doesn't sound like much, but 1/2" of dead air is still R-2.5. Insulation right up to the plates is easy to install, and that may offset any questionable and highly situation-dependent benefits of dead air.